4 ms·

Actually, there's been a few papers showing solid theoretical progress on understanding what structure in data deep networks are learning to represent. https:/

Actually, there's been a few papers showing solid theoretical progress on understanding what structure in data deep networks are learning to represent. https://github.com/gregversteeg/CorEx https://github.com/gregversteeg/CorEx http://arxiv.org/abs/1406.1222 http://arxiv.org/abs/1406.1222 http://arxiv.org/abs/1410.7404 http://arxiv.org/abs/1410.7404 Disclaimer: I reinvented the idea about a month or two ago, and ran into the existing papers when googling for joint entropy estimators after making some interesting graphs with hierarchical probabilistic programs. I only wish I could've been the first!

disgruntledphd2 11y ago
That's really interesting, especially with respect to the personality trait recovery. I'm particularly amazed given that I've never been able to replicate the Big5 model in its entirety against particular datasets. Only commenting on that portion, as it's what I know. The author doesn't appear to give any indication of which test was used, and where the sample was taken from. This makes me a little suspicious, but its probably just a disciplinary thing. Additionally, many tests reduce the pool of available questions for a scale after testing it against against other scales + theory. This means that what the author is recovering, is the scale imposed on the items by the original authors. Nonetheless, this has forced me to reinvestigate the Big5 model, so it's not all bad :). Massive appreciation for the links.
nl 11y ago
Possibly also of interest: Recursive Neural Networks Can Learn Logical Semantics[1] Probably not as theoretical as the work you referenced, but interesting to me because of the deeply practical outcomes in NLP. [1] http://www.aclweb.org/anthology/W/W15/W15-40.pdf#page=22 http://www.aclweb.org/anthology/W/W15/W15-40.pdf#page=22
j2kun 11y ago
I meant "deep nets" in terms of boolean circuits, which seems like a different beast from these papers.
Xcelerate 11y ago
Those papers are very interesting. I was particularly surprised that the method exactly recovers the Big-5 components of personality from the results of the survey questions. The whole idea of "simplest model that explains most of the data" has always been very appealing to me. The concept is closely tied to reproducing kernel Hilbert spaces, which have recently experienced a revival in interest due to the representer theorem.
- eli_gottlieb 11y ago
  >I was particularly surprised that the method exactly recovers the Big-5 components of personality from the results of the survey questions. Why? Is that a standard dataset? They could have just cherrypicked tasks. I also noticed that they seem to only have the actual computation working for discrete random variables (integer features) right now, which limits its applicability. They also seem to use mass-function estimators, which again can work well for discrete data while becoming statistically intractable when dealing with continuous random variables. >The whole idea of "simplest model that explains most of the data" has always been very appealing to me. In this case it's a bit more like, "The set of latent variables that best screen off the observables from each-other."
- emcq 11y ago
  You will probably be interested in Kolmogrov complexity, which studies this using an information theoretic lens. However it is just an abstract concept and not really something you can measure precisely.
blennon 11y ago
An interesting and early (1992) demonstration of deep nets learning latent structure in the visual domain was presented by Usui et al. They presented a deep autoencoder with the spectral values of color chips and show that the hidden layer representation corresponds to the Munsell color space where each dimension corresponds to a psychological/intuitive color attribute. Pretty amazing. Even cooler that they showed this in 1992. 10.1364/JOSAA.9.000516 Shiro Usui, Shigeki Nakauchi, and Masae Nakano, "Reconstruction of Munsell color space by a five-layer neural network," J. Opt. Soc. Am. A 9, 516-520 (1992)
auvrw 11y ago
> reinvented the idea about a month or two ago, and ran into the existing papers this happens all the time. i'm thankful that you took the time to post the results, because a discovery that's not communicated is not really a discovery. indeed, i suspect that what's going on in the field is as much a lack of communication as much as a lack of understanding. i wish the author of the original article had included links to at least a couple of of the "number of tools to probe the geometry of these hidden structures."

disgruntledphd2 11y ago

That's really interesting, especially with respect to the personality trait recovery. I'm particularly amazed given that I've never been able to replicate the Big5 model in its entirety against particular datasets. Only commenting on that portion, as it's what I know. The author doesn't appear to give any indication of which test was used, and where the sample was taken from. This makes me a little suspicious, but its probably just a disciplinary thing. Additionally, many tests reduce the pool of available questions for a scale after testing it against against other scales + theory. This means that what the author is recovering, is the scale imposed on the items by the original authors. Nonetheless, this has forced me to reinvestigate the Big5 model, so it's not all bad :). Massive appreciation for the links.

nl 11y ago

Possibly also of interest: Recursive Neural Networks Can Learn Logical Semantics[1] Probably not as theoretical as the work you referenced, but interesting to me because of the deeply practical outcomes in NLP. [1] http://www.aclweb.org/anthology/W/W15/W15-40.pdf#page=22 http://www.aclweb.org/anthology/W/W15/W15-40.pdf#page=22

j2kun 11y ago

I meant "deep nets" in terms of boolean circuits, which seems like a different beast from these papers.

Xcelerate 11y ago

Those papers are very interesting. I was particularly surprised that the method exactly recovers the Big-5 components of personality from the results of the survey questions. The whole idea of "simplest model that explains most of the data" has always been very appealing to me. The concept is closely tied to reproducing kernel Hilbert spaces, which have recently experienced a revival in interest due to the representer theorem.

eli_gottlieb 11y ago
>I was particularly surprised that the method exactly recovers the Big-5 components of personality from the results of the survey questions. Why? Is that a standard dataset? They could have just cherrypicked tasks. I also noticed that they seem to only have the actual computation working for discrete random variables (integer features) right now, which limits its applicability. They also seem to use mass-function estimators, which again can work well for discrete data while becoming statistically intractable when dealing with continuous random variables. >The whole idea of "simplest model that explains most of the data" has always been very appealing to me. In this case it's a bit more like, "The set of latent variables that best screen off the observables from each-other."
emcq 11y ago
You will probably be interested in Kolmogrov complexity, which studies this using an information theoretic lens. However it is just an abstract concept and not really something you can measure precisely.

blennon 11y ago

An interesting and early (1992) demonstration of deep nets learning latent structure in the visual domain was presented by Usui et al. They presented a deep autoencoder with the spectral values of color chips and show that the hidden layer representation corresponds to the Munsell color space where each dimension corresponds to a psychological/intuitive color attribute. Pretty amazing. Even cooler that they showed this in 1992. 10.1364/JOSAA.9.000516 Shiro Usui, Shigeki Nakauchi, and Masae Nakano, "Reconstruction of Munsell color space by a five-layer neural network," J. Opt. Soc. Am. A 9, 516-520 (1992)

auvrw 11y ago

> reinvented the idea about a month or two ago, and ran into the existing papers this happens all the time. i'm thankful that you took the time to post the results, because a discovery that's not communicated is not really a discovery. indeed, i suspect that what's going on in the field is as much a lack of communication as much as a lack of understanding. i wish the author of the original article had included links to at least a couple of of the "number of tools to probe the geometry of these hidden structures."